chain-of-verification
Draft → generate verification questions → answer independently via tools → revise. Catches hallucinated facts in reports and reviews. MANDATORY for Phase 4 security/test claims. Paper: Dhuliawala et al. 2023.
code-reviewer
8-aspect evidence-based code review/audit (security, correctness, compatibility, architecture, error handling, test gaps, type safety, simplicity) built on Google eng-practices, OWASP Top 10:2025, CWE Top 25, SemVer. Every finding needs file:line + a concrete failure scenario. Use when the user asks to review/audit code, check a PR, review before merge, in TDD Phase 4, via /check review or /run review.
conflict-detector
L1-L4 conflict detection between a proposed T4 task and pending-confirm sibling tasks. L1 (file overlap) + L2 (function/region overlap) ship in v3.7.0-beta.2 as deterministic bash. L3 (semantic LLM) + L4 (architectural LLM) are stubbed for v3.7.0-rc.1.
deep-debugging
Systematic debugging protocol for bugs that resist quick fixes. Use bisection, hypothesis trees, and scientific method when a bug isn't obvious from the stack trace. Goes beyond bugfix-quick for production-grade root cause analysis.
design-expert
UI/UX design expertise — component design, design system selection, responsive layout. Includes auto-detection from package.json and Context7 integration for library docs.
design-tokens
Generate a cohesive color-token system from ONE OKLCH brand hue — primary/secondary/background/foreground/muted/border with semantic light + dark, emitted as a Tailwind v4 @theme block or a CSS-variables fallback. Use when starting a design system, defining brand colors, setting up theming/dark-mode, or replacing scattered hardcoded colors with a derived palette.
design-vision-loop
Vision feedback loop for UI — render the running UI, screenshot it across viewports + dark mode, run deterministic gates then a multimodal critique against the design system, and iterate until it passes. Use after building or changing a UI component/page, when asked to 'check the UI', make it 'look right', or verify a build matches a design spec / Stitch screen.
documentation
Create Runbooks for operational documentation. For ADRs, tech specs, requirements, LLD and technical analysis use the tech-writing skill instead.
extension-detector
Detects when a new project-level skill / rule / command would reduce friction. Asks the user for confirmation before creating anything. NEVER writes plugin-level files — only `.claude/skills/`, `.claude/rules/`, `.claude/commands/` in the user's project.
failure-classifier
Classifies execution failures into F1-F5 (transient/local-logic/local-design/story-level/architectural). Outputs JSON with class, confidence, evidence, and recommended action. Deterministic rule-based classifier; no LLM call.
framework-expert
Unified framework expertise bundle. Lazy-loads relevant framework patterns (React, Vue, Angular, Next.js, Node.js, Python, Laravel, Go, Flutter, React Native, TypeScript) based on detected tech stack.
frontend-aesthetics
Anti-AI-slop taste layer for UI output — distinctive typography, dominant+accent color hierarchy, and implementation-matched polish. Use when building or restyling a UI/landing/component that should look designed rather than defaulted, or when the user asks for something 'beautiful', 'distinctive', 'less generic', or 'not another Tailwind template'.
git
Token-efficient git operations with security scanning and auto-split commits, plus auto-creating git worktrees to isolate workflow changes — merge on success, discard on failure.
mcp-security-auditor
On-demand audit of MCP usage. Reads the MCP audit log (mcp-audit.jsonl) written by the mcp-call-gate hook, surfaces blocked calls, rate-limit hits, suspicious input patterns. Companion to mcp-call-gate hook (which produces the audit log).
migration-helper
Guide safe database and code migrations with zero-downtime strategies.
monorepo
Detect and navigate monorepos correctly. Use when working with repos containing multiple packages (pnpm/yarn/npm workspaces, Turborepo, Nx, Lerna). Ensures commands run in correct package scope, dependencies are routed properly, and cross-package changes are coordinated.
motion-design
Restrained, accessible web motion — Framer Motion declarative patterns and CSS-only staggered reveals for high-impact moments, always bundle-aware and reduced-motion-safe. Use when adding animation/transitions to a UI (entrance reveals, list stagger, page/route transitions, micro-interactions) or when motion feels janky, heavy, or gratuitous.
perf-profiling
Systematic performance profiling and optimization across frontend (Core Web Vitals, code splitting, lazy loading), backend (N+1 queries, async), and database (EXPLAIN ANALYZE, indexing) layers. Use when the user reports slow code, latency, memory leaks, needs to benchmark, or wants to speed up an application. Measure first, optimize second. Applies Pareto principle — find the 20% of code causing 80% of slowness, fix that, not the rest.
permanent-memory-loader
Loads permanent_memory.md summary lines (≤120 always-loaded tokens, hard cap 200). Surfaces durable wisdom from past Epics into the current session. Silent if .claude/memory/ does not exist.
plan-archivist
Compresses a completed plan-tree branch into .claude/plans/archive/{NODE_ID}.summary.md. Removes the original node files (preserved by checkpoint history). Reduces always-loaded surface area as features ship.
plan-loader
Loads minimum hierarchical plan context (.claude/plans/) for current execution focus. Auto-invokes every session when plans exist. Stays under 800 always-loaded tokens regardless of plan tree size.
plan-orchestrator
Route hierarchical-planning intents to the correct backing script. Use when the user invokes /aura-frog:plan (with or without subcommand), mentions plan verbs (expand/next/replan/promote/archive/freeze/thaw/undo/status/conflicts), or types a plan-vocabulary bare word with .claude/plans/active.json present. Owns verb table, intent classifier, and 3-stage routing pipeline.
plan-validator
On-demand plan-tree validator. Runs all 8 invariants from spec §6.7 (parent existence, children integrity, no orphans, valid status, monotonic revision, test_ref existence, DAG no-cycles, freeze_reason). Refuses commits on violations.
preflight-validator
On-demand wrapper over scripts/preflight/run-all.sh. Runs Tier 1 bash linters against a file, command, or arbitrary content. Returns 0 pass / 1 warn / 2 fail. Used by /aura-frog:preflight check and contributors authoring custom hooks.
problem-solving
5 techniques for different problem types. Use when stuck or facing complex challenges.
project-context-loader
Load project conventions and generate session context. Use EXPLICITLY when needed, not automatically.
prompt-evaluator
Evaluate how you use Claude Code — analyze prompt patterns, feature utilization, and get improvement suggestions. Trigger: /prompts:evaluate, prompt analysis, usage evaluation, how am I using Claude
reasoning-trace-recorder
Auto-invokes for every Claude turn during active T4 task execution. Emits append-only trace events (file_read, output_claim, tool_call, decision) to .claude/plans/traces/{TASK_ID}.jsonl. Source of grounding-discipline checks. Disable: AF_TRACE_DISABLED=true.
refactor-expert
Guide safe, incremental refactoring that improves code quality without changing behavior.
run-orchestrator
Execute 5-phase TDD workflow for complex features. Use when the user invokes /run, asks to build/create/implement a feature, requests a complex multi-file change, or types 'fasttrack:'. Enforces phase gates, sprint contracts, and builder!=reviewer discipline.
scalable-thinking
Design for scale while keeping implementation simple (KISS).
self-consistency
Generate N independent reasoning paths and vote on the answer. Use for architectural trade-offs, ambiguous design decisions, or when single-path reasoning risks locking onto the first plausible answer. Paper: Wang et al. 2022.
self-healing-orchestrator
Proposes patches for F2 (local-logic) and F3 (local-design) failures. NEVER applies without user approval. Confidence ≥0.7 to propose; below that, escalates raw findings. Counts toward replan_budget. Per-task: max 1; per-session: max 5.
self-improve
Full learning loop for the Aura Frog plugin: analyze collected learning data from Supabase to identify success patterns, failure patterns, optimization opportunities, and agent performance trends, then apply learned improvements — update rules, adjust agent routing, modify workflow configurations, and generate knowledge base entries.
session-continuation
Manage workflow state across sessions with handoff and resume. TOON-based state persistence.
stitch-design
Generate UI designs using Google Stitch AI with optimized prompts
tech-writing
Write professional technical documents — requirements/PRD, tech spec/design doc, low-level design, decision records, and trade-off analysis — at a formality tier matched to the task. Grounds every claim in real code or a cited source, marks what it could not verify, and refuses to pad. Use when asked to write/update a spec, requirements, design doc, LLD, ADR, RFC, or technical analysis.
test-writer
Write tests with TDD following structured patterns. Ensures consistent AAA structure, proper coverage targets, and framework-specific conventions. Without this skill, tests lack consistent naming, miss coverage targets, and skip anti-pattern checks.
tree-of-thoughts
Structured reasoning — branch/evaluate/prune/expand search over solution space (tree), or a single ordered chain when the path is linear. Use for architecture with multi-step decisions, refactor planning, complex debug hypothesis trees, or step-by-step structured thinking. Papers: Yao et al. 2023 (ToT), sequential/chain-of-thought reasoning.
agent-detector
CRITICAL: MUST run for EVERY message. Detects agent, complexity, AND model automatically. Without this, tasks route to wrong agents and use wrong models, degrading quality and wasting tokens.
api-designer
Designs RESTful APIs with endpoint naming, versioning strategies (URL path, header-based), pagination (offset and cursor), error response schemas, and OpenAPI conventions. Use when the user asks about REST API design, creating endpoints, URL structure, API versioning, status codes, Swagger, or OpenAPI specs.
bugfix-quick
Fast bug fixes with root cause investigation + TDD. Enforces 'no fix without root cause' discipline and verification protocol. Without this skill, fixes are applied at symptoms instead of sources, and bugs return.
code-simplifier
Detect and simplify overly complex code. Apply KISS principle - less is more.